3634 papers
SRNeRV is a novel, parameter-efficient neural video representation framework that leverages scale self-similarity through a hybrid recursive sharing scheme to significantly reduce model size while achieving superior rate-distortion performance compared to traditional stacked multi-scale architectures.
GarmentPainter is an efficient framework that synthesizes high-fidelity, 3D-consistent garment textures in UV space by leveraging UV position maps for structural guidance and a type selection module for character-based control, all integrated into a standard diffusion model without architectural modifications.
This study leverages the MICCAI 2024 UWF4DR dataset to benchmark state-of-the-art deep learning models, including CNNs, Vision Transformers, and foundation models, in both spatial and frequency domains for image quality assessment, referable diabetic retinopathy detection, and diabetic macular edema identification using ultra-widefield imaging, demonstrating that feature-level fusion and frequency-domain representations yield robust and explainable results.
SiMO introduces a novel collaborative perception framework that utilizes Length-Adaptive Multi-Modal Fusion (LAMMA) and a "Pretrain-Align-Fuse-RD" training strategy to overcome sensor failures and semantic mismatches, ensuring robust performance across all individual modalities while maintaining effective multimodal integration.
This paper proposes a topologically stable variant of the Hough transform for line detection in point clouds, which replaces the traditional discretized voting scheme with a continuous score function and utilizes persistent homology to identify candidate lines via an efficient algorithm.
This paper introduces DynamicVGGT, a unified feed-forward framework that extends static 3D perception to dynamic 4D scene reconstruction for autonomous driving by jointly predicting current and future point maps, utilizing a Motion-aware Temporal Attention module for temporal coherence, and employing a Dynamic 3D Gaussian Splatting Head to explicitly model point motion and refine geometry.
The paper proposes WaDi, a novel one-step image synthesis framework that leverages the insight that weight direction changes are more critical than norm changes during distillation, introducing the parameter-efficient LoRaD adapter to achieve state-of-the-art performance with only 10% of trainable parameters.
This paper proposes a learning-free method for 6D object pose tracking that fuses event-based optical flow for high-speed pose propagation with a template-based correction strategy, demonstrating superior performance over state-of-the-art algorithms in highly dynamic scenarios where traditional RGB-D cameras struggle with motion blur and low frame rates.
This paper introduces a prototype-guided approach that leverages the intrinsic embedding geometry of diffusion models to identify and cluster concept representations, enabling the reliable erasure of broad, multi-faceted concepts while preserving overall image quality.
The paper proposes OSCAR, a label-free method that utilizes coupled latent spaces and neural implicit representations to accurately reconstruct complete 3D vertebral anatomy from partial ultrasound images by implicitly modeling acoustic shadowing and signal transmission, achieving an 80% improvement in HD95 score over state-of-the-art techniques.
The paper introduces SP-CLIP, a lightweight zero-shot action recognition framework that significantly improves performance on fine-grained and compositional actions by augmenting frozen vision-language models with structured, multi-level semantic prompts without requiring any additional parameter training or visual encoder modifications.
This paper proposes a retrieval-augmented framework for text-to-CT generation that leverages a 3D vision-language encoder to retrieve semantically related clinical cases and their anatomical annotations as structural proxies, thereby enhancing image fidelity and spatial controllability in a realistic inference setting without requiring ground-truth annotations.
This paper introduces a concept-guided fine-tuning framework that enhances Vision Transformer robustness against distribution shifts by automatically generating and aligning model attention with fine-grained semantic concepts rather than spurious background correlations.
HDR-NSFF is a novel framework that reconstructs dynamic high dynamic range (HDR) radiance fields from monocular videos by shifting from 2D pixel alignment to 4D spatio-temporal modeling, utilizing exposure-invariant motion estimation and generative priors to achieve state-of-the-art view synthesis while introducing the first real-world HDR-GoPro dataset for evaluation.
This paper introduces SlowBA, a novel backdoor attack against VLM-based GUI agents that utilizes a two-stage reward-level injection strategy and realistic pop-up triggers to induce excessive reasoning chains, thereby significantly increasing response latency while maintaining task accuracy and evading existing defenses.
This paper presents a large-scale comparative study using the Epic ReduAct dataset and over 3,000 human participants to demonstrate that while humans rely on sparse, semantically critical cues for egocentric action recognition, state-of-the-art AI models degrade more gradually by depending on contextual and low-level features, revealing fundamental divergences in how humans and machines process spatial and spatiotemporal information.
This paper introduces a label-free framework for evaluating Multiple Instance Learning heatmaps in histopathology, demonstrating through a large-scale benchmark that perturbation, LRP, and integrated gradients outperform attention-based methods, thereby enabling more reliable model validation and biological discovery.
The paper proposes SOT-GLP, a novel few-shot adaptation method for vision-language models that employs shared sparse optimal transport to partition visual regions among class-specific local prompts while maintaining global alignment, thereby achieving state-of-the-art performance in both classification accuracy and out-of-distribution detection by preserving the native feature geometry.
This paper introduces VLA, a prior-guided framework that enhances robotic manipulation by modeling discrete world-knowledge variations relative to an explicit current state prior, rather than predicting absolute future states, thereby achieving state-of-the-art performance and efficiency through its novel components: the Prior-Guided World Knowledge Extractor, Latent World Variation Quantization, and Conditional Variation Attention.
This paper introduces UniDiffDA, a unified analytical framework that decomposes diffusion-based data augmentation into three core components to enable a systematic, fair benchmarking of diverse methods across low-data classification tasks, ultimately offering practical design insights and reproducible code.